資產(chǎn)負債模型(Asset Liability Model)

1、w w w . w a t s o n w y a t t . c o m university of nsw 23 september 2002 guest lecture john ferraro james fitzpatrick agenda lasset liability modelling (alm) lworking at watson wyatt - james fitzpatrick - john ferraro please interrupt if you have a question the “problem” la trustees superannuation

2、fund has lost 10% over the last few years due to poor investment markets. with the coverage of assets over liabilities now at only 100%, further losses will mean that the fund will be under-funded. lthe “problem”: the trustee has asked for strategic advice to minimise the chance that the funds cover

3、age of liabilities falls further. the solution lthe trustee can benefit from an alm lwhat is an alm? lwhy is it needed here? - stochastic model versus deterministic - identifies risk lthe solution might mean a new strategy steps in alm lsetting asset assumptions lsetting asset allocation strategies

4、lbuilding the liability model lrunning simulations linterpreting results asset assumptions lglobal framework/model lreturns, standard deviations and correlations: - australian equities - international equities - australian fixed interest - listed property - direct property - international fixed inte

5、rest - cash - inflation, and more asset allocation strategies lthe strategy is the long term weight or allocation to each asset class lexample: australian equities 30% international equities 30% property 10% australian fixed interest 15% international fixed interest 15% liability model lactuarial in

6、put useful for superannuation funds limitate actuarial valuations in calculating contribution rates under each scenario lconflict: actuarial (prudent) versus alm (best estimate) assumptions running simulations lhow many is enough? 5000? lplatform? watson wyatt uses excel/visual basic ldata stored in

7、 access tables lcomplex funds can take a number of hours to run interpreting results lresults need to be interpreted and presented to the client lrecommendations may (and usually) do mean a change in strategy lexamples of results: example 1: return definition? vested benefits index xyz superannuatio

8、n fund vbi in year 2011 (active members only) (pension liability 80% immunised) 0% 50% 100% 150% 200% 0% growth 10% growth 20% growth 30% growth 40% growth 50% growth 60% growth 70% growth 80% growth 90% growth 100% growth 5th-25th percentile 25th-50th50th-75th example 2: risk definition? probabilit

9、y of under-funding xyz superannuation fund 80% immunised probability of vbi 100% in at least one year over the period - current 40% 50% 60% 70% 80% 90% 2002200320042005200620072008200920102011 year lgraduated in 2000 with a bachelor of commerce majoring in actuarial studies ljoined watson wyatt in 2

10、001 in the actuarial practice lwork in client teams helping trustees: - actuarial valuations - group life reviews - ad hoc problem solving (eg legislative changes) working at watson wyatt james fitzpatrick working at watson wyatt john ferraro ljoined watson wyatt in 1994 lworked in a number of departments: - it - it/administration - superannuation consulting - investment consulting lcurrent